mikelove

mikelove: Tshirt and data mob

2008-06-30T23:06:05.0000Z

I think the best thing to do is roll back influences if they don't affect the creative work of the influencee. There is a creative aspect to all the influence in the graph...scientists, writers, musicians, artists, etc. I think influence is observed in the works created by the influencee. I think the checking of influences to make sure it's more than "I feel inspired by this historical figure" needs to be done by a community, similar to Wikipedia's high bar for "encyclopedic content".

mikelove: Tshirt and data mob

2008-06-23T20:49:52.0000Z

Ae you going to set something uncomplicated up Mike?

mikelove: Tshirt and data mob

2008-06-22T16:46:09.0012Z

I agree that we do have to filter somehow, and hopefully in a way that is not too complicated. Maybe a 'notable person' attribute.

mikelove: Tshirt and data mob

2008-06-14T22:16:48.0000Z

Good morning

I just noticed the data mob around the Influence domain initiative, and I find it problematic for application like the influence viewer I wrote.

If half of the person in the graph become "basically unknown" person who feel inspired by Friedrich Nietzsche and alike then it will become quite pain browsing the graph in the search of historical figures and their influneces, since who want for example find me being influenced by Nagarjuna if the user is interested in the influences among Buddhist masters.

I would suggest to introduce some kind of flag attribute indicating which influence node was or is a historical or present figure with some level of publicity to be able to filter for this subgraph most people might be interested in.

The other maybe more clean idea which pops up in my mind would be to declare influence person to be a person who has gained some degree of publicity (or degree of more substantial influence or ...) either now or in the historical context and define influence as a relation between person in general (without having to be an influence node”

just me first ideas, curious what you think about this issue

cheers

martin

mikelove: Tshirt and data mob

2008-06-13T18:09:26.0007Z

Yeah, I agree re: significant influences but I'm not sure I agree about actual teachers/mentors if by that you mean people you've met personally. Take a look at mine and let me know what you think. I've listed a small set of people who I've never met but who have significantly affected my thinking, the way I live, or my creative expression.

mikelove: Tshirt and data mob

2008-06-12T20:35:58.0007Z

Sounds good. My only suggestion would be to have people list their actual teachers/mentors or otherwise people that significantly influenced their life work, as opposed to "I really enjoyed reading Jane Doe in college, so I will list her as an influence." What do you think?

mikelove: Tshirt and data mob

2008-06-12T18:39:27.0005Z

Hey, two things.

1) Do you have a Freebase tshirt yet? If not I'd love to send you one. Please let me know your size and postal address by email to kirrily@metaweb.com

2) I'm thinking of doing a data mob around the Influence domain, getting people to enter themselves and who they were influenced by, and then up the tree as far as they can manage. Just wanted to give you a heads-up.

Influence Node: Influences, positive and negative

2008-06-11T20:44:55.0022Z

So in the dataset I first uploaded, I included negative influence as just influence - partly to simplify the project and keep the barrier to entry low enough that it would actually get contributions. There are lots of degrees of influence, but I would err on the side of simplicity so it's easy for people to get going.

I think it's also fair to say David Hume influences Kant, Plato influences Aristotle, or Hegel influences Karl Marx, although the latter in the pair is in many ways rebelling against the core arguments of the former.

Peer Relationship: Problem Querying for Peers

2008-05-04T23:30:20.0007Z

James,

The thing at the end of a peers property is not a Person, and so that query finds no matches. If you query:

"peers":[{"type":null}]

you will see that the object there is one of these Peer Relationships with which this discussion is associated. To find the people in that relationship, you will need to do a query like:

"peers":[{"peers":[{

"type":"/people/person",

"name":null,

"date_of_birth":null

}]}]

That will find all of the participants in all of the peer relationships. However, as the discussion of spouse relationships on the Person type mentions, you will get Ezra Pound as a participant in each of these relationships, and will probably need to filter him out in whatever view you are creating.

Peer Relationship: Problem Querying for Peers

2008-05-04T22:31:56.0005Z

I get back 200 OK status but no data.

Sample query:

{
  "query" : [
    {
      "id" : null,
      "name" : "Ezra Pound",
      "peers" : [
        {
          "date_of_birth" : null,
          "id" : null,
          "limit" : 5,
          "name" : null,
          "type" : "/people/person"
        }
      ],
      "type" : "/influence/influence_node"
    }
  ]
}

Influence Node: Influences, positive and negative

2008-02-04T14:15:56.0012Z

This is a completely fascinating and powerful collaborative learning tool, and Mike deserves huge credit for getting the ball rolling.

I'd like to think aloud a bit about 'negative' influences, and see what anyone else thinks on this subject. It seems to me that when we refer to influences in a philosophical context, for example, it's generally assumed that we're talking about positive influences, ie.one or more aspects of a known party-of-the-first point of view that end up being utilised or subsumed or in some other fashion became incorporated into the expressed thought of the party-of-the-second. Correct?

However, negative influences can be equally important, in that critical examination and subsequent rejection on stated or unstated grounds very often plays a vital role in determining the subsequent direction of an influencee's thinking and conclusions. These negative influences can be well known, buried in footnotes, or, sometimes, inferred with some degree of confidence.

So, my question is, do negative influences legitimately belong in Mikes influence type, as conceived by him and as utilized by this type's rapid expansion in use? Or does 'influence' require two explicit types, positive and negative? Any thoughts?

Influence Node: influence vs. refers to (it would make sense to reverse the link direction)

2007-08-23T10:03:39.0005Z

The concepts of 'school' or 'tradition' own a decisive advantage over the concept of influence, as they do not imply a direction of the link.
Infuence implies that a dead person, say Leonardo da Vinci, actively acts on a living person, say Andy Warhol.
In reality the relation points to the past in all cases. Andy Warhol 'refers to' Leonardo da Vinci instead of being activly 'influenced' by the dead artist.
This distinction between 'influence' and different forms of 'refering to' as well as it's implications have been discussed very strikingly by Michael Baxandall in his 'Excursion against Influence', which can be found in his 1985 book 'Patterns of Intention' pp. 58-61.
Functionally the difference between 'influencing' and 'refering to' is equivalent to the difference between 'bibliometric coupling' and 'co-citation' in the field of Bibliometrics. Whereas 'influence' implies a static actor in the past, modern day 'reference' alters our image of the past actor.
A classic example would be 9/11: By refering to films-scenes where high-rise buildings are burning or destroyed, the terrorists changed our perception of films like 'The Fight Club' or 'Die Hard' forever. By 'refering to' the films they changed our perception of the past. On the other hand it would not make sense to sue one of the directors for 'influencing' the terrorists.

Influence Node: School/Tradition from Wikipedia

2007-08-22T16:57:36.0005Z

I tried do adopt the school/tradition field from Wikipedia by creating my own type. I hope this might be usefull and will be used in the genealogy!